# minuteTemp > Live, minute-level temperature data for the weather stations that settle daily and hourly temperature prediction markets on Kalshi, Polymarket, Polymarket US, Robinhood, IBKR (ForecastEx) and Coinbase, with 20 forecast models, live odds, and REST and WebSocket APIs. 73 cities. minuteTemp is for prediction-market traders and trading bots. It shows each market's settlement station as the market reads it, the running high and low, forecasts, and how each platform settles, so traders can see where a market is heading before the official number is published. ## Coverage - 73 cities and 78 settlement stations; 23 stations report every minute. - Kalshi: 24 cities with daily high and low markets, and 4 hourly markets on the Kalshi Weather Index. Coinbase lists the same Kalshi markets. - Polymarket: 53 cities. Polymarket US: 6 cities. - ForecastEx contracts on Robinhood (23 cities) and IBKR (31 cities). ## Pricing - Free ($0): For casual weather watchers and lead-gen exploration. 3-min delayed UI updates, 1 day of history. - UI ($14.99/mo): For fast, ad-free dashboards without external API complexity. 1-min UI updates, 3 days of history. - Starter ($29.99/mo): For individual traders who want API access and live workflows. 1-min UI updates, API 15 req/min, 7 days of history. - Pro ($89.99/mo): For serious API users, automation, and team workflows. 1-min UI updates, API 100 req/min, 30 days of history. ## Settlement rules (as of 2026-10-19) - Kalshi: Daily high and low settle on the station's NWS climate record (the CLI report) as published by The Weather Company, then the National Weather Service: whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). Settles once the value is published, usually 3 to 12 hours after midnight. - Kalshi hourly: Settles on the Kalshi Weather Index, a weighted index of several airport stations, not one station: the last official index minute at or before the top of the hour, looking back at most 60 minutes, to 0.01 °F. If the hour has no official minute, Kalshi's market outcome review decides it. - Coinbase: Lists Kalshi's daily and hourly markets under the same tickers and order book; they settle by Kalshi's rules, the daily markets on the station's climate record and the hourly markets on the Kalshi Weather Index. - Polymarket: Settles on NOAA's weather.gov timeseries for the station: the highest (or lowest) reading of the local calendar day, from the hourly data in whole °F for US stations and every reading in whole °C elsewhere. Weather Underground's Daily Observations table is the fallback, and the source for Taipei, Jinan and Zhengzhou (whole °C). - Polymarket US: Settles on the NWS Daily Climate Report (CLI) for the station: whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). Usually the next morning, around 8 AM ET. - ForecastEx: Daily high (UH) and low (UL) contracts settle on Weather Underground's Daily Observations table for one airport station: the highest or lowest reading of the local calendar day, whole °F, with strikes as thresholds 1 °F apart. International daily highs (SH) use the same table in whole °C and have no low. - Robinhood: A broker listing ForecastEx contracts, not Kalshi's: the daily UH/UL contracts settle by the ForecastEx rule. Hourly HRU contracts settle each hour T, Central Time, on the last Weather Underground observation after T minus 60 minutes and up to T, whole °F. - IBKR: Lists the ForecastEx daily contracts, including the international SH highs, plus Kalshi daily highs in 7 cities, which settle by Kalshi's rules. Each city page (/market/) states the station, source, rounding and day window per platform. ## API - [API overview](https://minutetemp.com/api.md): base URL, authentication, rate limits, every endpoint and WebSocket channel. - [OpenAPI spec (JSON)](https://api.minutetemp.com/v1/openapi.json): REST API under https://api.minutetemp.com/api/v1/, API key in the X-API-Key header. Also as [YAML](https://api.minutetemp.com/v1/openapi.yaml). - [AsyncAPI spec](https://api.minutetemp.com/v1/asyncapi.yaml): WebSocket channels /ws/api/1m and /ws/api/5m; get a ticket from POST /api/v1/ws-ticket. ## Docs - [Market Settlement](https://minutetemp.com/docs/market-settlement.md): How temperature prediction markets settle — data sources, timing, and resolution rules for Kalshi, Polymarket, Polymarket US, and the ForecastEx contracts on Robinhood and IBKR. - [How Kalshi Hourly Temperature Markets Settle: The Kalshi Weather Index Explained](https://minutetemp.com/docs/kalshi-weather-index.md): How Kalshi hourly temperature markets settle on the Kalshi Weather Index: member stations, weights, offsets, quorum, official minutes and the settlement minute. - [Temperature Data Sources](https://minutetemp.com/docs/temperature-data.md): How minuteTemp collects and processes real-time temperature observations from METAR, NWS, and Weather Underground. - [NWS Data](https://minutetemp.com/docs/nws-data.md): How minuteTemp integrates National Weather Service forecast and observation data. - [Forecast Models](https://minutetemp.com/docs/forecast-models.md): Overview of the 20 forecast models compared on minuteTemp — GFS, HRRR, NAM, ECMWF, and more. - [Oracle Scores](https://minutetemp.com/docs/oracle-scores.md): How minuteTemp Oracle Scores work — methodology for ranking forecast model accuracy using MAE, bias, and ASOS ground truth. - [Reading the Dashboard](https://minutetemp.com/docs/reading-dashboard.md): How to read and interpret the minuteTemp dashboard — charts, indicators, and data panels explained. ## Key pages - [Live Temperature Data for Kalshi Hourly & Daily Markets](https://minutetemp.com): Live data for 73 cities and Kalshi hourly temperature markets: 1-minute station data, the Kalshi Weather Index every minute, 20 forecast models and live odds. - [Today's High & Low Temperatures for Kalshi & Polymarket](https://minutetemp.com/today): Live highs and lows so far today with forecast model consensus and spread for all 73 cities with Kalshi, Polymarket, Polymarket US, Robinhood and IBKR temperature markets. - [Markets](https://minutetemp.com/markets): Every temperature market we track: 73 cities on Kalshi, Polymarket, Polymarket US, Robinhood and IBKR, with live settlement-station data, forecasts and each station's data resolution. - [Kalshi Weather Markets: Temperature Settlement & Stations](https://minutetemp.com/kalshi): Kalshi daily and hourly temperature markets: the station behind each of 24 cities, what the rules settle on, and live data for every market. - [Kalshi Hourly Temperature Markets & Weather Index, Live](https://minutetemp.com/kalshi/hourly): Live Kalshi hourly temperature markets for Miami, New York City, Chicago and Los Angeles Coastal: the Kalshi Weather Index every minute, stations and strikes. - [Polymarket Temperature Markets: Event Slugs & Stations](https://minutetemp.com/polymarket): Polymarket daily highest and lowest temperature markets in 53 cities: each city's event slug, the station it settles on, and live data for every market. - [Coinbase Weather Markets: Temperature Tickers & Stations](https://minutetemp.com/coinbase): Coinbase temperature prediction markets are Kalshi's: 48 daily high and low series in 24 cities and 4 hourly series, with tickers, stations and live data. - [Robinhood Temperature Markets: UH/UL Tickers, Stations & Settlement](https://minutetemp.com/robinhood): Robinhood temperature markets are ForecastEx contracts: daily high (UH) and low (UL) in 23 cities, hourly (HRU) in 5, settled on Weather Underground. Tickers and stations. - [Robinhood Hourly Temperature Markets: HRULGA, HRULAX & More](https://minutetemp.com/robinhood/hourly): Robinhood hourly temperature contracts (ForecastEx HRULGA, HRULAX, HRUSFO, HRUMIA and HRULAS): live strikes, MinuteTemp's reading at the settling station and how each hour settles. - [IBKR Temperature Markets: ForecastEx UH/UL & Kalshi Highs](https://minutetemp.com/ibkr): Interactive Brokers temperature markets: ForecastEx daily high (UH) and low (UL) in 24 cities on Weather Underground, plus Kalshi daily highs in 7. Tickers and stations. - [Pricing](https://minutetemp.com/pricing): Compare minuteTemp plans: free delayed data, live Kalshi hourly markets on every paid plan from $14.99/mo, and Starter and Pro API plans from ~21,600 requests/day. - [Documentation](https://minutetemp.com/docs): Guides and documentation for minuteTemp — temperature data sources, forecast models, market settlement, and API reference. ## Optional - [Full text](https://minutetemp.com/llms-full.txt): this file, every city's settlement rules and the docs above in full. - [Sitemap](https://minutetemp.com/sitemap.xml) --- # Settlement by city (as of 2026-10-19) ## Atlanta (https://minutetemp.com/market/atl) - Kalshi: Kalshi's Atlanta daily high and low markets settle on the NWS climate report (CLI) for Hartsfield-Jackson Atlanta (KATL), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Atlanta daily markets (KXHIGHTATL and KXLOWTATL) under the same tickers; they settle by Kalshi's rules on KATL. - Polymarket: Polymarket's Atlanta daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Hartsfield-Jackson Atlanta (KATL): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Atlanta daily high and low (UHATL and ULATL) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Hartsfield-Jackson Atlanta (KATL): whole °F, the highest or lowest reading of the local calendar day. ## Austin (https://minutetemp.com/market/aus) - Kalshi: Kalshi's Austin daily high and low markets settle on the NWS climate report (CLI) for Austin-Bergstrom International (KAUS), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Austin daily markets (KXHIGHAUS and KXLOWTAUS) under the same tickers; they settle by Kalshi's rules on KAUS. - IBKR (Kalshi): IBKR also lists Kalshi's Austin daily high (KXHIGHAUS), which settles by Kalshi's rules on KAUS. - Polymarket: Polymarket's Austin daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Austin-Bergstrom International (KAUS): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Austin daily high and low (UHAUS and ULAUS) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Austin-Bergstrom International (KAUS): whole °F, the highest or lowest reading of the local calendar day. ## Boston (https://minutetemp.com/market/bos) - Kalshi: Kalshi's Boston daily high and low markets settle on the NWS climate report (CLI) for Boston Logan International (KBOS), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Boston daily markets (KXHIGHTBOS and KXLOWTBOS) under the same tickers; they settle by Kalshi's rules on KBOS. - Robinhood and IBKR (ForecastEx): ForecastEx's Boston daily high and low (UHBOS and ULBOS) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Boston Logan International (KBOS): whole °F, the highest or lowest reading of the local calendar day. ## Charlotte (https://minutetemp.com/market/clt) - Robinhood and IBKR (ForecastEx): ForecastEx's Charlotte daily high and low (UHCLT and ULCLT) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Charlotte Douglas International (KCLT): whole °F, the highest or lowest reading of the local calendar day. ## Chicago (https://minutetemp.com/market/chi) - Kalshi: Kalshi's Chicago daily high and low markets settle on the NWS climate report (CLI) for Chicago Midway (KMDW), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Kalshi hourly: Kalshi's hourly Chicago market (KXTEMPCHIHS) settles on the Kalshi Weather Index, not one station: the last official index minute at or before the top of the hour, looking back at most 60 minutes, to 0.01 °F. If the hour has no official minute, Kalshi's market outcome review decides it. - Coinbase: Coinbase lists Kalshi's Chicago markets under the same tickers and settles them by Kalshi's rules: the daily markets (KXHIGHCHI and KXLOWTCHI) on KMDW, and the hourly market (KXTEMPCHIHS) on the Kalshi Weather Index. - IBKR (Kalshi): IBKR also lists Kalshi's Chicago daily high (KXHIGHCHI), which settles by Kalshi's rules on KMDW. - Polymarket: Polymarket's Chicago daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Chicago O'Hare International (KORD): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Polymarket US: Polymarket US's Chicago daily high and low markets settle on the NWS Daily Climate Report (CLI) for Chicago Midway (KMDW): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's Chicago daily high and low (UHMDW and ULMDW) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Chicago Midway (KMDW): whole °F, the highest or lowest reading of the local calendar day. ## Dallas (https://minutetemp.com/market/dal) - Kalshi: Kalshi's Dallas daily high and low markets settle on the NWS climate report (CLI) for Dallas/Fort Worth International (KDFW), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Dallas daily markets (KXHIGHTDAL and KXLOWTDAL) under the same tickers; they settle by Kalshi's rules on KDFW. - Polymarket: Polymarket's Dallas daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Dallas Love Field (KDAL): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Dallas daily high and low (UHDFW and ULDFW) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Dallas/Fort Worth International (KDFW): whole °F, the highest or lowest reading of the local calendar day. ## Denver (https://minutetemp.com/market/den) - Kalshi: Kalshi's Denver daily high and low markets settle on the NWS climate report (CLI) for Denver International (KDEN), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Denver daily markets (KXHIGHDEN and KXLOWTDEN) under the same tickers; they settle by Kalshi's rules on KDEN. - IBKR (Kalshi): IBKR also lists Kalshi's Denver daily high (KXHIGHDEN), which settles by Kalshi's rules on KDEN. - Polymarket: Polymarket's Denver daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Buckley Space Force Base (KBKF): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Denver daily high and low (UHBKF and ULBKF) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Buckley Space Force Base (KBKF): whole °F, the highest or lowest reading of the local calendar day. ## Detroit (https://minutetemp.com/market/dtw) - Robinhood and IBKR (ForecastEx): ForecastEx's Detroit daily high and low (UHDTW and ULDTW) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Detroit Metropolitan Wayne County (KDTW): whole °F, the highest or lowest reading of the local calendar day. ## Houston (https://minutetemp.com/market/hou) - Kalshi: Kalshi's Houston daily high and low markets settle on the NWS climate report (CLI) for Houston Hobby (KHOU), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Houston daily markets (KXHIGHTHOU and KXLOWTHOU) under the same tickers; they settle by Kalshi's rules on KHOU. - IBKR (Kalshi): IBKR also lists Kalshi's Houston daily high (KXHIGHTHOU), which settles by Kalshi's rules on KHOU. - Polymarket: Polymarket's Houston daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Houston Hobby (KHOU): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Houston daily high and low (UHHOU and ULHOU) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Houston Hobby (KHOU): whole °F, the highest or lowest reading of the local calendar day. ## Jacksonville (https://minutetemp.com/market/jax) - Robinhood and IBKR (ForecastEx): ForecastEx's Jacksonville daily high and low (UHJAX and ULJAX) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Jacksonville International (KJAX): whole °F, the highest or lowest reading of the local calendar day. ## Las Vegas (https://minutetemp.com/market/las) - Kalshi: Kalshi's Las Vegas daily high and low markets settle on the NWS climate report (CLI) for Las Vegas Harry Reid International (KLAS), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Las Vegas daily markets (KXHIGHTLV and KXLOWTLV) under the same tickers; they settle by Kalshi's rules on KLAS. - Robinhood and IBKR (ForecastEx): ForecastEx's Las Vegas daily high and low (UHLAS and ULLAS) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Las Vegas Harry Reid International (KLAS): whole °F, the highest or lowest reading of the local calendar day. - Robinhood hourly: Robinhood's hourly Las Vegas contracts (HRULAS) settle each hour T, Central Time, on the last Weather Underground observation for KLAS after T minus 60 minutes and up to T: whole °F. ## Los Angeles (https://minutetemp.com/market/la) - Kalshi: Kalshi's Los Angeles daily high and low markets settle on the NWS climate report (CLI) for Los Angeles International (KLAX), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Kalshi hourly: Kalshi's hourly Los Angeles market (KXTEMPLAXHS) settles on the Kalshi Weather Index, not one station: the last official index minute at or before the top of the hour, looking back at most 60 minutes, to 0.01 °F. If the hour has no official minute, Kalshi's market outcome review decides it. - Coinbase: Coinbase lists Kalshi's Los Angeles markets under the same tickers and settles them by Kalshi's rules: the daily markets (KXHIGHLAX and KXLOWTLAX) on KLAX, and the hourly market (KXTEMPLAXHS) on the Kalshi Weather Index. - IBKR (Kalshi): IBKR also lists Kalshi's Los Angeles daily high (KXHIGHLAX), which settles by Kalshi's rules on KLAX. - Polymarket: Polymarket's Los Angeles daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Los Angeles International (KLAX): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Polymarket US: Polymarket US's Los Angeles daily high markets settle on the NWS Daily Climate Report (CLI) for Los Angeles International (KLAX): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's Los Angeles daily high and low (UHLAX and ULLAX) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Los Angeles International (KLAX): whole °F, the highest or lowest reading of the local calendar day. - Robinhood hourly: Robinhood's hourly Los Angeles contracts (HRULAX) settle each hour T, Central Time, on the last Weather Underground observation for KLAX after T minus 60 minutes and up to T: whole °F. ## Miami (https://minutetemp.com/market/mia) - Kalshi: Kalshi's Miami daily high and low markets settle on the NWS climate report (CLI) for Miami International (KMIA), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Kalshi hourly: Kalshi's hourly Miami market (KXTEMPMIAH) settles on the Kalshi Weather Index, not one station: the last official index minute at or before the top of the hour, looking back at most 60 minutes, to 0.01 °F. If the hour has no official minute, Kalshi's market outcome review decides it. - Coinbase: Coinbase lists Kalshi's Miami markets under the same tickers and settles them by Kalshi's rules: the daily markets (KXHIGHMIA and KXLOWTMIA) on KMIA, and the hourly market (KXTEMPMIAH) on the Kalshi Weather Index. - IBKR (Kalshi): IBKR also lists Kalshi's Miami daily high (KXHIGHMIA), which settles by Kalshi's rules on KMIA. - Polymarket: Polymarket's Miami daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Miami International (KMIA): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Polymarket US: Polymarket US's Miami daily high markets settle on the NWS Daily Climate Report (CLI) for Miami International (KMIA): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's Miami daily high and low (UHMIA and ULMIA) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Miami International (KMIA): whole °F, the highest or lowest reading of the local calendar day. - Robinhood hourly: Robinhood's hourly Miami contracts (HRUMIA) settle each hour T, Central Time, on the last Weather Underground observation for KMIA after T minus 60 minutes and up to T: whole °F. ## Minneapolis (https://minutetemp.com/market/msp) - Kalshi: Kalshi's Minneapolis daily high and low markets settle on the NWS climate report (CLI) for Minneapolis-Saint Paul International (KMSP), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Minneapolis daily markets (KXHIGHTMIN and KXLOWTMIN) under the same tickers; they settle by Kalshi's rules on KMSP. - Robinhood and IBKR (ForecastEx): ForecastEx's Minneapolis daily high and low (UHMSP and ULMSP) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Minneapolis-Saint Paul International (KMSP): whole °F, the highest or lowest reading of the local calendar day. ## Nashville (https://minutetemp.com/market/bna) - Robinhood and IBKR (ForecastEx): ForecastEx's Nashville daily high and low (UHBNA and ULBNA) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Nashville International (KBNA): whole °F, the highest or lowest reading of the local calendar day. ## New Orleans (https://minutetemp.com/market/msy) - Kalshi: Kalshi's New Orleans daily high and low markets settle on the NWS climate report (CLI) for New Orleans Louis Armstrong International (KMSY), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's New Orleans daily markets (KXHIGHTNOLA and KXLOWTNOLA) under the same tickers; they settle by Kalshi's rules on KMSY. - IBKR (ForecastEx): ForecastEx's New Orleans daily high and low (UHMSY and ULMSY) on IBKR settle on Weather Underground's Daily Observations table for New Orleans Louis Armstrong International (KMSY): whole °F, the highest or lowest reading of the local calendar day. ## New York (https://minutetemp.com/market/nyc) - Kalshi: Kalshi's New York daily high and low markets settle on the NWS climate report (CLI) for New York City Central Park (KNYC), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Kalshi hourly: Kalshi's hourly New York market (KXTEMPNYCHS) settles on the Kalshi Weather Index, not one station: the last official index minute at or before the top of the hour, looking back at most 60 minutes, to 0.01 °F. If the hour has no official minute, Kalshi's market outcome review decides it. - Coinbase: Coinbase lists Kalshi's New York markets under the same tickers and settles them by Kalshi's rules: the daily markets (KXHIGHNY and KXLOWTNYC) on KNYC, and the hourly market (KXTEMPNYCHS) on the Kalshi Weather Index. - IBKR (Kalshi): IBKR also lists Kalshi's New York daily high (KXHIGHNY), which settles by Kalshi's rules on KNYC. - Polymarket: Polymarket's New York daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for LaGuardia Airport (KLGA): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Polymarket US: Polymarket US's New York daily high and low markets settle on the NWS Daily Climate Report (CLI) for New York City Central Park (KNYC): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's New York daily high and low (UHLGA and ULLGA) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for LaGuardia Airport (KLGA): whole °F, the highest or lowest reading of the local calendar day. - Robinhood hourly: Robinhood's hourly New York contracts (HRULGA) settle each hour T, Central Time, on the last Weather Underground observation for KLGA after T minus 60 minutes and up to T: whole °F. ## Oklahoma City (https://minutetemp.com/market/okc) - Kalshi: Kalshi's Oklahoma City daily high and low markets settle on the NWS climate report (CLI) for Will Rogers World Airport (KOKC), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Oklahoma City daily markets (KXHIGHTOKC and KXLOWTOKC) under the same tickers; they settle by Kalshi's rules on KOKC. - Robinhood and IBKR (ForecastEx): ForecastEx's Oklahoma City daily high and low (UHOKC and ULOKC) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Will Rogers World Airport (KOKC): whole °F, the highest or lowest reading of the local calendar day. ## Philadelphia (https://minutetemp.com/market/phl) - Kalshi: Kalshi's Philadelphia daily high and low markets settle on the NWS climate report (CLI) for Philadelphia International (KPHL), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Philadelphia daily markets (KXHIGHPHIL and KXLOWTPHIL) under the same tickers; they settle by Kalshi's rules on KPHL. - Polymarket US: Polymarket US's Philadelphia daily high markets settle on the NWS Daily Climate Report (CLI) for Philadelphia International (KPHL): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's Philadelphia daily high and low (UHPHL and ULPHL) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Philadelphia International (KPHL): whole °F, the highest or lowest reading of the local calendar day. ## Phoenix (https://minutetemp.com/market/phx) - Kalshi: Kalshi's Phoenix daily high and low markets settle on the NWS climate report (CLI) for Phoenix Sky Harbor International (KPHX), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Phoenix daily markets (KXHIGHTPHX and KXLOWTPHX) under the same tickers; they settle by Kalshi's rules on KPHX. - Robinhood and IBKR (ForecastEx): ForecastEx's Phoenix daily high and low (UHPHX and ULPHX) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Phoenix Sky Harbor International (KPHX): whole °F, the highest or lowest reading of the local calendar day. ## San Antonio (https://minutetemp.com/market/sat) - Kalshi: Kalshi's San Antonio daily high and low markets settle on the NWS climate report (CLI) for San Antonio International (KSAT), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's San Antonio daily markets (KXHIGHTSATX and KXLOWTSATX) under the same tickers; they settle by Kalshi's rules on KSAT. - Robinhood and IBKR (ForecastEx): ForecastEx's San Antonio daily high and low (UHSAT and ULSAT) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for San Antonio International (KSAT): whole °F, the highest or lowest reading of the local calendar day. ## San Diego (https://minutetemp.com/market/san) - Kalshi: Kalshi's San Diego daily high and low markets settle on the NWS climate report (CLI) for San Diego International (KSAN), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's San Diego daily markets (KXHIGHTSAN and KXLOWTSAN) under the same tickers; they settle by Kalshi's rules on KSAN. ## San Francisco (https://minutetemp.com/market/sfo) - Kalshi: Kalshi's San Francisco daily high and low markets settle on the NWS climate report (CLI) for San Francisco International (KSFO), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's San Francisco daily markets (KXHIGHTSFO and KXLOWTSFO) under the same tickers; they settle by Kalshi's rules on KSFO. - Polymarket: Polymarket's San Francisco daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for San Francisco International (KSFO): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Polymarket US: Polymarket US's San Francisco daily high markets settle on the NWS Daily Climate Report (CLI) for San Francisco International (KSFO): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Robinhood and IBKR (ForecastEx): ForecastEx's San Francisco daily high and low (UHSFO and ULSFO) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for San Francisco International (KSFO): whole °F, the highest or lowest reading of the local calendar day. - Robinhood hourly: Robinhood's hourly San Francisco contracts (HRUSFO) settle each hour T, Central Time, on the last Weather Underground observation for KSFO after T minus 60 minutes and up to T: whole °F. ## Seattle (https://minutetemp.com/market/sea) - Kalshi: Kalshi's Seattle daily high and low markets settle on the NWS climate report (CLI) for Seattle-Tacoma International (KSEA), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Seattle daily markets (KXHIGHTSEA and KXLOWTSEA) under the same tickers; they settle by Kalshi's rules on KSEA. - Polymarket: Polymarket's Seattle daily high and low markets settle on the hourly data of NOAA's weather.gov timeseries for Seattle-Tacoma International (KSEA): whole °F, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - Robinhood and IBKR (ForecastEx): ForecastEx's Seattle daily high and low (UHSEA and ULSEA) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Seattle-Tacoma International (KSEA): whole °F, the highest or lowest reading of the local calendar day. ## Washington D.C. (https://minutetemp.com/market/dca) - Kalshi: Kalshi's Washington D.C. daily high and low markets settle on the NWS climate report (CLI) for Ronald Reagan Washington National (KDCA), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Washington D.C. daily markets (KXHIGHTDC and KXLOWTDC) under the same tickers; they settle by Kalshi's rules on KDCA. - Robinhood and IBKR (ForecastEx): ForecastEx's Washington D.C. daily high and low (UHDCA and ULDCA) on Robinhood and IBKR settle on Weather Underground's Daily Observations table for Ronald Reagan Washington National (KDCA): whole °F, the highest or lowest reading of the local calendar day. ## Amsterdam (https://minutetemp.com/market/amsterdam) - Polymarket: Polymarket's Amsterdam daily high and low markets settle on NOAA's weather.gov timeseries for Amsterdam Schiphol Airport (EHAM): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Ankara (https://minutetemp.com/market/ankara) - Polymarket: Polymarket's Ankara daily high and low markets settle on NOAA's weather.gov timeseries for Ankara Esenboga International Airport (LTAC): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Auckland (https://minutetemp.com/market/auckland) - Polymarket: Polymarket's Auckland daily high markets settle on NOAA's weather.gov timeseries for Auckland International Airport (NZAA): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Beijing (https://minutetemp.com/market/beijing) - Polymarket: Polymarket's Beijing daily high and low markets settle on NOAA's weather.gov timeseries for Beijing Capital International Airport (ZBAA): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Buenos Aires (https://minutetemp.com/market/buenos-aires) - Polymarket: Polymarket's Buenos Aires daily high and low markets settle on NOAA's weather.gov timeseries for Buenos Aires Ministro Pistarini International Airport (SAEZ): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Busan (https://minutetemp.com/market/busan) - Polymarket: Polymarket's Busan daily high and low markets settle on NOAA's weather.gov timeseries for Busan Gimhae International Airport (RKPK): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Cape Town (https://minutetemp.com/market/cape-town) - Polymarket: Polymarket's Cape Town daily high and low markets settle on NOAA's weather.gov timeseries for Cape Town International Airport (FACT): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Chengdu (https://minutetemp.com/market/chengdu) - Polymarket: Polymarket's Chengdu daily high and low markets settle on NOAA's weather.gov timeseries for Chengdu Shuangliu International Airport (ZUUU): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Chongqing (https://minutetemp.com/market/chongqing) - Polymarket: Polymarket's Chongqing daily high and low markets settle on NOAA's weather.gov timeseries for Chongqing Jiangbei International Airport (ZUCK): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Frankfurt (https://minutetemp.com/market/frankfurt) - IBKR (ForecastEx): ForecastEx's Frankfurt daily high (SHDDF) on IBKR settles on Weather Underground's Daily Observations table for Frankfurt am Main Airport (EDDF): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Helsinki (https://minutetemp.com/market/helsinki) - Polymarket: Polymarket's Helsinki daily high and low markets settle on NOAA's weather.gov timeseries for Helsinki Vantaa Airport (EFHK): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Istanbul (https://minutetemp.com/market/istanbul) - Polymarket: Polymarket's Istanbul daily high and low markets settle on NOAA's weather.gov timeseries for Istanbul Airport (LTFM): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Jakarta (https://minutetemp.com/market/jakarta) - Polymarket: Polymarket's Jakarta daily high markets settle on NOAA's weather.gov timeseries for Jakarta Halim Perdanakusuma International Airport (WIHH): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Jeddah (https://minutetemp.com/market/jeddah) - Polymarket: Polymarket's Jeddah daily high and low markets settle on NOAA's weather.gov timeseries for Jeddah King Abdulaziz International Airport (OEJN): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Kuala Lumpur (https://minutetemp.com/market/kuala-lumpur) - Polymarket: Polymarket's Kuala Lumpur daily high and low markets settle on NOAA's weather.gov timeseries for Kuala Lumpur International Airport (WMKK): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Lagos (https://minutetemp.com/market/lagos) - Polymarket: Polymarket's Lagos daily high markets settle on NOAA's weather.gov timeseries for Lagos Murtala Muhammed International Airport (DNMM): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## London (https://minutetemp.com/market/london) - Polymarket: Polymarket's London daily high and low markets settle on NOAA's weather.gov timeseries for London City Airport (EGLC): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Lucknow (https://minutetemp.com/market/lucknow) - Polymarket: Polymarket's Lucknow daily high and low markets settle on NOAA's weather.gov timeseries for Chaudhary Charan Singh International Airport (VILK): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Madrid (https://minutetemp.com/market/madrid) - Polymarket: Polymarket's Madrid daily high and low markets settle on NOAA's weather.gov timeseries for Adolfo Suarez Madrid-Barajas Airport (LEMD): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Mexico City (https://minutetemp.com/market/mexico-city) - Polymarket: Polymarket's Mexico City daily high and low markets settle on NOAA's weather.gov timeseries for Mexico City Benito Juarez International Airport (MMMX): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Milan (https://minutetemp.com/market/milan) - Polymarket: Polymarket's Milan daily high and low markets settle on NOAA's weather.gov timeseries for Milan Malpensa Airport (LIMC): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Montreal (https://minutetemp.com/market/montreal) - IBKR (ForecastEx): ForecastEx's Montreal daily high (SHYUL) on IBKR settles on Weather Underground's Daily Observations table for Montreal Trudeau International Airport (CYUL): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Moscow (https://minutetemp.com/market/moscow) - Polymarket: Polymarket's Moscow daily high and low markets settle on NOAA's weather.gov timeseries for Moscow Vnukovo Airport (UUWW): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Munich (https://minutetemp.com/market/munich) - Polymarket: Polymarket's Munich daily high and low markets settle on NOAA's weather.gov timeseries for Munich Franz Josef Strauss International Airport (EDDM): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Panama City (https://minutetemp.com/market/panama-city) - Polymarket: Polymarket's Panama City daily high and low markets settle on NOAA's weather.gov timeseries for Panama City Marcos A. Gelabert International Airport (MPMG): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Paris (https://minutetemp.com/market/paris) - Polymarket: Polymarket's Paris daily high and low markets settle on NOAA's weather.gov timeseries for Paris Le Bourget Airport (LFPB): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - IBKR (ForecastEx): ForecastEx's Paris daily high (SHFPO) on IBKR settles on Weather Underground's Daily Observations table for Paris Orly Airport (LFPO): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## São Paulo (https://minutetemp.com/market/saopaulo) - Polymarket: Polymarket's São Paulo daily high and low markets settle on NOAA's weather.gov timeseries for São Paulo Guarulhos International Airport (SBGR): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Seoul (https://minutetemp.com/market/seoul) - Polymarket: Polymarket's Seoul daily high and low markets settle on NOAA's weather.gov timeseries for Incheon International Airport (RKSI): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Shenzhen (https://minutetemp.com/market/shenzhen) - Polymarket: Polymarket's Shenzhen daily high and low markets settle on NOAA's weather.gov timeseries for Shenzhen Bao'an International Airport (ZGSZ): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Singapore (https://minutetemp.com/market/singapore) - Polymarket: Polymarket's Singapore daily high and low markets settle on NOAA's weather.gov timeseries for Singapore Changi Airport (WSSS): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - IBKR (ForecastEx): ForecastEx's Singapore daily high (SHSSS) on IBKR settles on Weather Underground's Daily Observations table for Singapore Changi Airport (WSSS): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Sydney (https://minutetemp.com/market/sydney) - IBKR (ForecastEx): ForecastEx's Sydney daily high (SHSSY) on IBKR settles on Weather Underground's Daily Observations table for Sydney Kingsford Smith Airport (YSSY): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Taipei (https://minutetemp.com/market/taipei) - Polymarket: Polymarket's Taipei daily high and low markets settle on Weather Underground's Daily Observations table for Taipei Songshan Airport (RCSS): whole °C, the highest (or lowest) reading of the local calendar day. ## Tel Aviv (https://minutetemp.com/market/tel-aviv) - Polymarket: Polymarket's Tel Aviv daily high and low markets settle on NOAA's weather.gov timeseries for Tel Aviv Ben Gurion Airport (LLBG): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Tokyo (https://minutetemp.com/market/tokyo) - Polymarket: Polymarket's Tokyo daily high and low markets settle on NOAA's weather.gov timeseries for Tokyo Haneda Airport (RJTT): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - IBKR (ForecastEx): ForecastEx's Tokyo daily high (SHJTT) on IBKR settles on Weather Underground's Daily Observations table for Tokyo Haneda Airport (RJTT): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Toronto (https://minutetemp.com/market/toronto) - Polymarket: Polymarket's Toronto daily high and low markets settle on NOAA's weather.gov timeseries for Toronto Pearson International Airport (CYYZ): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. - IBKR (ForecastEx): ForecastEx's Toronto daily high (SHYYZ) on IBKR settles on Weather Underground's Daily Observations table for Toronto Pearson International Airport (CYYZ): whole °C, the highest reading of the local calendar day. There is no daily low contract. ## Warsaw (https://minutetemp.com/market/warsaw) - Polymarket: Polymarket's Warsaw daily high and low markets settle on NOAA's weather.gov timeseries for Warsaw Chopin Airport (EPWA): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Wellington (https://minutetemp.com/market/wellington) - Polymarket: Polymarket's Wellington daily high and low markets settle on NOAA's weather.gov timeseries for Wellington International Airport (NZWN): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Wuhan (https://minutetemp.com/market/wuhan) - Polymarket: Polymarket's Wuhan daily high and low markets settle on NOAA's weather.gov timeseries for Wuhan Tianhe International Airport (ZHHH): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Louisville (https://minutetemp.com/market/sdf) - Kalshi: Kalshi's Louisville daily high and low markets settle on the NWS climate report (CLI) for Louisville Muhammad Ali International (KSDF), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Louisville daily markets (KXHIGHTSDF and KXLOWTSDF) under the same tickers; they settle by Kalshi's rules on KSDF. ## Newark (https://minutetemp.com/market/ewr) - Kalshi: Kalshi's Newark daily high and low markets settle on the NWS climate report (CLI) for Newark Liberty International (KEWR), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Newark daily markets (KXHIGHTEWR and KXLOWTEWR) under the same tickers; they settle by Kalshi's rules on KEWR. ## Trenton (https://minutetemp.com/market/ttn) - Kalshi: Kalshi's Trenton daily high and low markets settle on the NWS climate report (CLI) for Trenton Mercer (KTTN), as published by The Weather Company (the NWS if it is unavailable): whole °F, the NWS climate day, midnight to midnight local standard time (1 AM to 1 AM during daylight saving time). - Coinbase: Coinbase lists Kalshi's Trenton daily markets (KXHIGHTTTN and KXLOWTTTN) under the same tickers; they settle by Kalshi's rules on KTTN. ## Guangzhou (https://minutetemp.com/market/guangzhou) - Polymarket: Polymarket's Guangzhou daily high and low markets settle on NOAA's weather.gov timeseries for Guangzhou Baiyun International Airport (ZGGG): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Jinan (https://minutetemp.com/market/jinan) - Polymarket: Polymarket's Jinan daily high and low markets settle on Weather Underground's Daily Observations table for Jinan Yaoqiang International Airport (ZSJN): whole °C, the highest (or lowest) reading of the local calendar day. ## Karachi (https://minutetemp.com/market/karachi) - Polymarket: Polymarket's Karachi daily high and low markets settle on NOAA's weather.gov timeseries for Karachi Jinnah International Airport (OPKC): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Manila (https://minutetemp.com/market/manila) - Polymarket: Polymarket's Manila daily high and low markets settle on NOAA's weather.gov timeseries for Manila Ninoy Aquino International Airport (RPLL): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Qingdao (https://minutetemp.com/market/qingdao) - Polymarket: Polymarket's Qingdao daily high and low markets settle on NOAA's weather.gov timeseries for Qingdao Jiaodong International Airport (ZSQD): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Shanghai (https://minutetemp.com/market/shanghai) - Polymarket: Polymarket's Shanghai daily high and low markets settle on NOAA's weather.gov timeseries for Shanghai Pudong International Airport (ZSPD): whole °C, the highest (or lowest) reading of the local calendar day, with Weather Underground as the fallback. ## Zhengzhou (https://minutetemp.com/market/zhengzhou) - Polymarket: Polymarket's Zhengzhou daily high and low markets settle on Weather Underground's Daily Observations table for Zhengzhou Xinzheng International Airport (ZHCC): whole °C, the highest (or lowest) reading of the local calendar day. --- Source: https://minutetemp.com/docs/market-settlement # How Temperature Markets Settle Temperature markets don't settle on what your weather app shows. They settle on an official record, such as the NWS Climatological Report (CLI) or a station's Weather Underground history — and the path from observation to settlement has important implications. ## The CLI Report The NWS Climatological Report (CLI) is the official daily weather summary for a station. It includes the day's official high temperature, low temperature, and average — calculated from raw sensor data (one-minute observations) without the additional rounding layers that affect publicly displayed temperatures. Critically, CLI data **has undergone NWS review**. Faulty sensor readings are adjusted or removed. Temporary station outages are noted. This is what makes it the authoritative record. The CLI is why you might watch a temperature all day and see 70°F displayed, but the official record settles at 69°F or 71°F. The displayed value went through the [rounding pipeline](https://minutetemp.com/docs/temperature-data); the CLI used raw sensor data. ## Kalshi Settlement Kalshi's daily temperature rules name The Weather Company first, then the National Weather Service, as the source of the official high or low for the market's NWS climate location (for example CLINYC for Central Park). The rules cite the value as published on The Weather Company's Kalshi data portal (weather.com/kalshi), and the first non-preliminary value counts. That official value is the station's NWS daily climate record, which minuteTemp tracks through the CLI report. Kalshi settles a market once the official value is published, so there is no fixed settlement window: most daily markets settle roughly 3 to 12 hours after they close at midnight, and some not until the early afternoon ET. minuteTemp records settlement outcomes from Kalshi's API, tracking which markets resolved and at what value. For example, if a Kalshi market asks "Will NYC's high be above 75°F?" and the official high is 74°F, the "No" contracts pay out — regardless of whether real-time observations briefly showed 75°F or 76°F during the day. See [Kalshi weather markets](https://minutetemp.com/kalshi) for every city's station and [Kalshi hourly temperature markets](https://minutetemp.com/kalshi/hourly) for the hourly markets. ## Coinbase Settlement Coinbase lists Kalshi's temperature markets, operated by Kalshi: the same tickers, order book and stations. Settlement follows Kalshi's rules for each market, so a market settles the same way on Coinbase as on Kalshi. Coinbase's prediction markets are not available to Nevada or Michigan residents. See [Coinbase weather markets](https://minutetemp.com/coinbase) for the series it lists and [Coinbase hourly temperature markets](https://minutetemp.com/coinbase/hourly) for the hourly ones. ## Polymarket Settlement Polymarket (polymarket.com, the international exchange) temperature markets do not use the CLI. For most cities their rules name NOAA's weather.gov timeseries for the market's station as the resolution source: a daily high resolves on the highest “Temp” reading of the local calendar day, and a daily low on the lowest. US markets use the hourly data (“Show Hourly Data”) in whole °F; other cities use the station's readings in whole °C. Weather Underground's Daily Observations table is used only if NOAA's data for the day is unavailable by 11:59 PM ET the next day. Some cities also use a different station than Kalshi, such as LaGuardia (KLGA) for New York and O'Hare (KORD) for Chicago. A few cities differ. Taipei (RCSS) settles on Weather Underground's Daily Observations table in whole °C, not NOAA. Hong Kong settles on the Hong Kong Observatory's Daily Extract to 0.1 °C; minuteTemp does not cover it. On the dashboard, minuteTemp's Polymarket estimate follows the same rule: for US stations, the highest or lowest routine hourly METAR, rounded to whole °F as NOAA's table shows it; for other NOAA cities, every METAR in whole °C; for Taipei, the same METARs, which are the rows of Weather Underground's table. NOAA's hourly view also lists some special reports (SPECI), so on a rare day the official value can differ from the estimate. The rules can vary by market, so check each market's resolution criteria on Polymarket for the exact data source. See [Polymarket temperature markets](https://minutetemp.com/polymarket) for every city's event slug and station. ## Polymarket US Settlement Polymarket US is a separate, CFTC-regulated US exchange with its own markets and order books. Its daily high markets (New York, Chicago, Miami, Los Angeles, San Francisco and Philadelphia) and daily low markets (New York and Chicago) settle on the National Weather Service's Daily Climate Report (CLI) for Central Park (KNYC), Midway (KMDW), KMIA, KLAX, KSFO and KPHL. Kalshi uses the same stations, but its rules name The Weather Company first and the NWS second; Polymarket US names the NWS report itself. Because the source is the CLI, the day is the NWS climate day (midnight to midnight local standard time), the same as Kalshi. Settlement is normally the next morning, around 8am ET. ## Robinhood and IBKR (ForecastEx) Settlement Robinhood's temperature markets are ForecastEx contracts: ForecastEx is the exchange, and Robinhood is a broker, so they are not Kalshi contracts. Interactive Brokers lists the same ForecastEx contracts, and also Kalshi's daily high for Austin, Chicago, Denver, Houston, Los Angeles, Miami and New York. The rules below apply to its ForecastEx contracts only; its Kalshi contracts follow Kalshi's rules and station, and minuteTemp shows them under Kalshi. **Daily high and low** (UH and UL products, such as UHLGA and ULLGA) settle on the **Weather Underground Daily Observations table** for one airport station: the maximum temperature shown for the local calendar day for highs, the minimum for lows. The day is final once the first observation of the next day is published, and settlement is normally no later than the day after expiration. Strikes are thresholds 1 °F apart rather than brackets: high contracts ask whether the high will _exceed_ a strike (“exceed 85 F”), low contracts whether the low will be _below_ one (“below 52 F”). The station can differ from Kalshi's: New York uses LaGuardia (KLGA), not Central Park (KNYC), and Denver uses Buckley Space Force Base (KBKF), not Denver International (KDEN). **International daily high** (SH products, on IBKR: Frankfurt EDDF, Montreal CYUL, Paris Orly LFPO, Singapore WSSS, Sydney YSSY, Tokyo Haneda RJTT and Toronto CYYZ) settle the same way, in **whole °C** as the metric Daily Observations table shows them, not converted from °F. Strikes are 1 °C apart (“exceed 15 C”), and there is no low. minuteTemp does not cover Vancouver's SH product yet: it settles on Vancouver Harbour (CYHC), not Vancouver International. **Hourly** (HRU products, on Robinhood for a few cities): each hour T, Central Time, settles on the last Weather Underground observation for the station after T − 60 minutes and up to T, in whole °F, with the NWS time series as the fallback. The hour is final once Weather Underground publishes the first observation after T. See [Robinhood weather markets](https://minutetemp.com/robinhood) for every city's products and station, [Robinhood hourly temperature markets](https://minutetemp.com/robinhood/hourly) for the HRU contracts and [IBKR temperature markets](https://minutetemp.com/ibkr) for what Interactive Brokers lists. ## Timeline: Observation to Settlement 1 Throughout the day Real-time observations reported every 5–60 minutes (preliminary) 2 End of observation day ASOS stations compile daily summary from one-minute data 3 Evening / overnight NWS reviews data, applies quality control corrections 4 Next morning The station's official high/low for the NWS climate day is published: by The Weather Company, and by NWS in the CLI (Climatological Report) 5 Hours later, at no fixed time Kalshi settles on The Weather Company's value (NWS if it is unavailable); other venues settle on their own sources ## When CLI Differs From What You Saw It's common for the CLI to report a high or low that differs from what you tracked in real-time. This happens because: - •**Rounding differences:** Real-time data goes through the F→C→F rounding pipeline; CLI uses raw sensor values - •**Sensor corrections:** NWS may remove outlier readings from a malfunctioning sensor - •**Averaging differences:** CLI uses 2-minute averaged one-minute data; public feeds may use instantaneous or differently averaged values ## Edge Cases - •**Station outages:** If a station goes offline during the observation period, the CLI may note missing data. Market resolution rules vary by platform for these scenarios. - •**Corrected readings:** Occasionally the NWS corrects a CLI after initial publication. Some platforms have provisions for amended reports. - •**Multi-station cities:** Different stations may report different temperatures. Know which station the market references — e.g., Kalshi's NYC markets reference Central Park (KNYC), while Polymarket and the ForecastEx contracts on Robinhood and IBKR use LaGuardia (KLGA). Warning The temperature you watched all day may not be the number that settles the market. The CLI undergoes NWS review and uses different data processing than real-time feeds. See [understanding temperature data](https://minutetemp.com/docs/temperature-data) for how observation values can differ from official records. [NWS Data Sources](https://minutetemp.com/docs/nws-data)[Forecast Models](https://minutetemp.com/docs/forecast-models) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/kalshi-weather-index # How Kalshi Hourly Temperature Markets Settle Kalshi's hourly temperature markets don't settle on one thermometer. They settle on the Kalshi Weather Index, a number built every minute from several airport stations. This is how that number is made and which minute decides the hour. ## What the Kalshi Weather Index is The Kalshi Weather Index is a temperature for a metro area, published by Kalshi once a minute in Fahrenheit to two decimal places. It is a weighted average of several airport stations. Kalshi runs the index; the station readings are 1-minute airport observations supplied by Synoptic Data. Kalshi publishes the index for 13 areas. 4 of them have an hourly market today: Miami (KXTEMPMIAH), New York City (KXTEMPNYCHS), Chicago (KXTEMPCHIHS), Los Angeles Coastal (KXTEMPLAXHS). See every city and its live page on [Kalshi hourly temperature markets](https://minutetemp.com/kalshi/hourly). ## Member stations and weights Each index has a list of member stations with equal base weights. A station that does not report is not counted as zero. The index is the weighted average of the stations that were accepted for that minute. Index Member stations A minute needs [Miami](https://minutetemp.com/hourly/miami?platform=kalshi) KMIA, KOPF, KFLL, KFXE, KPMP at least 4 of 5 stations and 80% of the base weight [New York City](https://minutetemp.com/hourly/nyc?platform=kalshi) KLGA, KEWR, KTEB, KCDW, KHPN, KFRG, KSMQ, KISP at least 7 of 8 stations and 87.5% of the base weight [Chicago](https://minutetemp.com/hourly/chicago?platform=kalshi) KMDW, KORD, KPWK, KDPA, KARR at least 4 of 5 stations and 80% of the base weight [Los Angeles Coastal](https://minutetemp.com/hourly/la-coastal?platform=kalshi) KHHR, KLAX, KSMO, KLGB, KSNA at least 4 of 5 stations and 80% of the base weight [Houston](https://minutetemp.com/hourly/houston?platform=kalshi) KHOU, KIAH, KSGR, KDWH, KCXO at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Dallas-Fort Worth](https://minutetemp.com/hourly/dfw?platform=kalshi) KDFW, KDAL, KAFW, KDTO, KRBD at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [San Francisco Bay Area](https://minutetemp.com/hourly/sf-bay?platform=kalshi) KSFO, KOAK, KSJC, KHWD, KCCR, KLVK, KAPC at least 6 of 7 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Puget Sound](https://minutetemp.com/hourly/puget-sound?platform=kalshi) KSEA, KBFI, KPAE, KTIW, KOLM at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Philadelphia Delaware Valley](https://minutetemp.com/hourly/phl-delaware-valley?platform=kalshi) KPHL, KPNE, KTTN, KILG, KMIV at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Greater Boston](https://minutetemp.com/hourly/greater-boston?platform=kalshi) KBOS, KBED, KORH, KOWD, KTAN at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Southeast Michigan](https://minutetemp.com/hourly/southeast-michigan?platform=kalshi) KDTW, KDET, KFNT, KPTK, KYIP at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Kansas City](https://minutetemp.com/hourly/kansas-city?platform=kalshi) KMCI, KMKC, KOJC, KIXD, KLXT at least 4 of 5 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) [Minneapolis-St. Paul](https://minutetemp.com/hourly/minneapolis-st-paul?platform=kalshi) KMSP, KMIC, KSTP, KFCM at least 3 of 4 stations and 80% of the base weight (our default; Kalshi hasn't published this city's rule) ## Offsets and weekly calibration Every member station has an offset, and each city has a reference value, the weighted average of its stations' offsets. Before averaging, each reading has its station's offset subtracted and the city reference added back. When every station reports, the offsets cancel out and the index equals the plain weighted average of the readings. They matter when a station is missing: every remaining reading has already been adjusted toward the city level, so a warm or cool station dropping out moves the index less. Kalshi re-estimates the offsets weekly. Base weights do not change with a calibration. The calculation runs in Celsius and is converted to Fahrenheit and rounded to 0.01°F at the end. ## Provisional, official, degraded, missing - **Official.** A reading has to reach Kalshi within 5 minutes of its minute. Kalshi then publishes one value for that minute and never revises it. Only official minutes settle a market. - **Provisional.** Before that, minuteTemp shows its own value computed from the readings received so far, about 2 minutes into the minute. It can still change. - **Degraded.** An official minute where a station was missing, used a fallback reading or had a quality-control substitute. It still counts for settlement. - **Missing.** If too few stations are accepted, the minute has no official value. The table above gives each city's minimum. Fallback readings include the station's own METAR report, at most 75 minutes old. ## Which minute settles the hour - The market settles on the **last official index minute at or before the top of the hour**. The top-of-hour minute itself counts. - If that minute is missing, the rule looks back to the newest official minute, at most 60 minutes earlier. - Because a minute turns official 5 minutes after it ends, the settlement value is known about 5 minutes after the hour. - If the hour has no official minute at all, Kalshi's market outcome review (Rule 7.1) applies. Trader Takeaway The hour does not settle on its high, its low or its average. One minute decides it, so the last few minutes before the hour matter more than the rest. ## Reading the market - Strikes are one degree apart and read “75° or above”. That strike resolves Yes when the settlement value is 75.00° or higher. - Each series has one open hour at a time, for the coming top of the hour. The strike ladder moves with the temperature from hour to hour. - The hour in a ticker is Eastern time for every city. A Los Angeles market ending in 16 settles at 4 PM Eastern, which is 1 PM in Los Angeles. ## How minuteTemp tracks it Each [hourly index page](https://minutetemp.com/kalshi/hourly) rebuilds the index every minute and shows the official value beside it, every member station's reading and its pull on the index, and an HRRR 15-minute forecast adjusted to the live index. Past hours are listed with the value we computed. The 4 cities with a Kalshi hourly market (Miami, New York City, Chicago and Los Angeles Coastal) also show the live strike ladder, and each past hour is checked against Kalshi's final value for the market. The other 9 cities have no market, so they have no ladder and their hourly settlements are informational. Kalshi's daily high and low markets work differently: one station per city. See [Kalshi weather and temperature markets](https://minutetemp.com/kalshi). [How Markets Settle](https://minutetemp.com/docs/market-settlement)[Kalshi hourly markets, live](https://minutetemp.com/kalshi/hourly) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/temperature-data # Understanding Temperature Data Why the temperature you see on your screen may not match the official record — and why that matters for trading. ## The Rounding Problem Official ASOS (Automated Surface Observing System) temperatures are 2-minute averages of sensor readings, reported in whole-degree Fahrenheit. This is the standard used for official weather records and market settlement. However, the publicly available 5-minute data that most weather apps and websites display goes through a multi-step conversion that introduces rounding errors: 1. The sensor reads a precise Fahrenheit value (e.g., 77.6°F) 2. It gets converted to Celsius (25.33°C) 3. Rounded to the nearest whole Celsius (25°C) 4. Converted back to Fahrenheit (77.0°F) 5. Rounded to whole Fahrenheit (77°F) The original 77.6°F reading could end up displayed as 77°F or 78°F depending on exactly where the rounding falls. That 1-2°F difference can determine whether a market contract pays out. Example Actual sensor reading: **77.6°F** After rounding pipeline: **77°F** (displayed on weather apps) Official CLI report: **78°F** (uses raw sensor data with proper averaging) ## Why Displayed Temps Differ From Official Records Most weather websites and apps source their data from ASOS stations, but the data they receive often lacks the proper 2-minute averaging required for official records. Some sources pull from whole-degree Celsius values and convert back to Fahrenheit, compounding the rounding error. No known near-real-time public source combines correct 2-minute averaging with proper Fahrenheit conversion. This means every real-time temperature you see online is an approximation — not the value that will appear in the official record. The Iowa Environmental Mesonet (IEM) deliberately marks some 5-minute near-real-time temperatures as missing specifically because they don't meet the official 2-minute averaging standard. They'd rather show no data than misleading data. ## Hourly vs 5-Minute Data Not all observation intervals are equal in precision: - •**5-minute intervals** — Most frequent but least precise. Subject to the full rounding pipeline described above. Good for tracking trends, unreliable for exact values. - •**Hourly observations (XX:51 to XX:54)** — Official hourly readings with less rounding ambiguity. Celsius values are reported to one decimal place, reducing conversion error. - •**SPECI reports** — Special observations triggered by significant weather changes. These contain exact temperature readings without the standard 5-minute rounding ambiguity. ## Top-of-the-Hour Unreliability You might expect the "top of the hour" temperature to be measured at exactly :00. In reality, hourly reports arrive several minutes early due to collection procedures dating back to the 1970s. ASOS sites have reset times between :50 and :59 after each hour. If the temperature changes by 1°F between the reset time and the actual top of the hour, that change may go unreported. The METAR T-group temperatures can also differ from mandatory integer Celsius values, creating additional discrepancies. For traders watching a temperature approach a strike price in the final minutes before the hour, this timing gap matters. ## Why The Observation Line Can Differ minuteTemp does not treat every temperature-shaped data point the same way. The main observation line is built from non-report sensor observations first, so it preserves a consistent real-time station trend instead of mixing in every later bulletin or promoted correction. That means a METAR T-group, CLI, or DSM-derived temperature can still matter for official interpretation without becoming part of the primary observation line. If a report-derived point lands at a time when the canonical sensor observation is missing, the chart may visually bridge across that minute using the surrounding non-report observations. - •**Observation line:** optimized to show the live sensor timeline traders are watching minute to minute. - •**Report-derived values:** can still influence official context such as settlement interpretation or promoted highs and lows, even when they are not drawn as a normal line point. - •**Fallback behavior:** if a report-derived value is all we have, it can still be surfaced as the latest available reading rather than dropped entirely. In practice, this means a chart line, a latest displayed value, and the eventual official record can briefly disagree without any one of them being "wrong." They are answering slightly different questions about the same station. ## How minuteTemp Represents Precision minuteTemp treats each observed temperature as a range rather than a single exact value. Because many public readings originate from integer Celsius, each reading implies a Fahrenheit window that is roughly 1.8°F wide. - •**Chart overlay:** red line = estimated max, blue line = estimated min - •**Tooltip labels:** hover shows Actual, Min, and Max together - •**External payloads:** REST and WebSocket observations include`temp_min_f`and`temp_max_f` This makes uncertainty explicit in both the UI and the API, so internal and external consumers can reason about the same bounds. Trader Takeaway Don't trust displayed temperatures within 1-2°F of a strike price. The official CLI number — which is what markets settle on — could differ from what you're seeing in real-time. See [how temperature markets settle](https://minutetemp.com/docs/market-settlement) for details on the official record. Further Reading - [Iowa Mesonet: ASOS Temperature Standards](https://mesonet.agron.iastate.edu/onsite/news.phtml?id=1469) - [NWS: Automated Surface Observing System (ASOS)](https://www.weather.gov/asos/) [NWS Data Sources](https://minutetemp.com/docs/nws-data) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/nws-data # NWS Data Sources Explained The National Weather Service delivers temperature data through multiple formats with very different precision levels. Knowing which is which matters for trading. ## Data Precision Hierarchy NWS temperature data comes in 5 tiers, from least to most precise. For market trading, only the bottom tier (CLI) is what ultimately determines settlement: Low1. 5-Minute Intervals Most frequent but least precise. Subject to the full F→C→round→F rounding pipeline. Good for tracking trends, unreliable for exact settlement values. Medium2. Hourly Readings (XX:51–XX:54) Official hourly observations with Celsius values rounded to one decimal place. Less rounding ambiguity than 5-minute data. High3. SPECI Reports Special observations triggered by significant weather changes (e.g., thunderstorms, rapid temp shifts). Contain exact temperature readings without standard rounding. High4. DSM (Daily Summary Message) Intra-day summary reports with very high precision. Published during the day, providing early looks at daily extremes. Official5. CLI (Climatological Report) The gold standard. Official highs, lows, and averages calculated from raw sensor data without additional rounding layers. Undergoes NWS review. This is what markets settle on. ## One-Minute Observations (OMO) ASOS stations continuously capture minute-by-minute temperature data. These One-Minute Observations exist but are not publicly displayed on the standard NWS timeseries pages. OMO data is used internally to calculate 6-hour and 24-hour maximums and minimums, which feed into the official daily summary. So while you can't see the one-minute readings, they influence the final CLI values that settle markets. ## Preliminary vs Final Data Every temperature observation you see in real-time is **preliminary**. It has not undergone NWS quality control review. This means: - •Faulty sensor readings might be adjusted or removed in the final record - •Temporary sensor malfunctions during the day could produce outlier readings - •The NWS may apply corrections before publishing the official CLI report The CLI report — typically published the following morning — is the first dataset that has undergone NWS review and represents the official record. ## What minuteTemp Shows minuteTemp displays real NWS station observations polled every 60 seconds. When a station goes quiet, charts may visually bridge short gaps as display-only synthetic points while the backend keeps stored observations limited to source readings. On the chart, observed readings include a precision envelope: a red line for the estimated maximum and a blue line for the estimated minimum. Hover tooltips show**Actual**, **Min**, and **Max** values directly. External API and WebSocket observation payloads include these bounds as`temp_min_f`and`temp_max_f`so downstream systems can use the same uncertainty model. Warning Real-time observations are preliminary and subject to change. Only the CLI report is final and NWS-reviewed. See [how markets settle](https://minutetemp.com/docs/market-settlement) for the timeline from observation to official record. Further Reading - [NWS: Understanding Climate Observations](https://www.weather.gov/bgm/climatology_obs) - [Iowa Mesonet: ASOS Temperature Standards](https://mesonet.agron.iastate.edu/onsite/news.phtml?id=1469) [Temperature Data](https://minutetemp.com/docs/temperature-data)[Market Settlement](https://minutetemp.com/docs/market-settlement) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/forecast-models # Weather Forecast Models Guide minuteTemp tracks 20 weather models from agencies around the world. Here's what they are and how they differ. ## What Is a Forecast Model? Numerical Weather Prediction (NWP) models are complex computer simulations that use atmospheric physics equations across three-dimensional grids to predict future weather. Each model divides the atmosphere into a grid of cells and simulates how temperature, pressure, moisture, and wind evolve over time. Different models use different grid sizes (resolution), cover different geographic areas, and update at different intervals. Higher resolution models capture local effects better but can only forecast a shorter time horizon. Lower resolution models can see further into the future but miss local details. ## US-Focused Models (NOAA/NCEP) These models are produced by NOAA/NCEP and focus on US weather prediction: Model Resolution Horizon Updates HRRR (High-Resolution Rapid Refresh) 3km 48 hours Every hour NBM (National Blend of Models) 3km 10 days Every hour GFS (Global Forecast System) 25km 16 days Every 6 hours NAM (North American Mesoscale) 5km 84 hours Every 6 hours **HRRR:** Highest resolution US model. Updates hourly with a 48-hour horizon. US only. **NBM:** Bias-corrected blend of multiple models. Applies statistical post-processing to raw model output. **GFS:** Primary US global model. Covers the full 16-day forecast horizon at 25km resolution. **NAM:** Regional model covering North America. 5km resolution captures more local terrain effects than GFS (25km). ## Global Models (ECMWF) ECMWF operates two global forecast systems: ECMWF IFS25km · 10 days · Every 6 hours Operated by the European Centre. Frequently cited in verification studies as a top-performing global model. ECMWF AIFS25km · 15 days · Every 6 hours AI-enhanced version of IFS. Combines traditional physics with machine learning. Note: ECMWF data via Open-Meteo may have a slight delay compared to direct ECMWF access. ## AI/ML & International Models minuteTemp also tracks AI models and forecasts from weather agencies around the world, providing independent perspectives: Model Provider Resolution Notes AIGFS NOAA/NCEP 25km NOAA's AI/ML global model. Different strengths than traditional models, especially for extreme events. GEM Canada (ECCC) 10–25km Useful for cross-border weather patterns affecting northern US cities. ICON Germany (DWD) 13km Strong European model with global coverage. ARPEGE France (Meteo France) 25km French global model with 4-day horizon. JMA GSM Japan (JMA) 55km Coarser resolution but provides an independent global perspective. UKMO UK (Met Office) 10km High-resolution Unified Model. 7-day horizon. GRAPES China (CMA) 25km Chinese global model. 10-day horizon. ACCESS Australia (BOM) 40km Australian global model. Useful for southern hemisphere perspective. ## Models by Forecast Horizon Different models cover different time ranges. Here's how they compare: 0–6 hours HRRR Highest resolution (3km), hourly updates capture rapidly evolving conditions. 6–48 hours HRRR + NBM HRRR provides 3km detail; NBM adds statistical bias correction. 2–7 days NBM + ECMWF IFS NBM is bias-corrected for US stations; ECMWF provides an independent global perspective. 7+ days GFS + ECMWF IFS These models extend to 10-16 days. Forecast skill decreases significantly beyond day 7. ## Model Consensus When 10 or more models agree on a temperature within a narrow range, that's a strong signal of confidence. minuteTemp highlights model consensus on the dashboard. Conversely, when models diverge significantly — especially when HRRR and ECMWF disagree by more than 2°F — uncertainty is high. This doesn't mean the forecast is wrong; it means the atmosphere is in a state where small differences in initial conditions lead to very different outcomes. See [reading the dashboard](https://minutetemp.com/docs/reading-dashboard) for how to spot consensus and divergence visually. ## Resolution Explained Model resolution describes the size of each grid cell. Smaller cells = more detail: - •**3km (~2 miles)** — Neighborhood-level. Can capture sea breezes, urban heat islands, terrain effects. - •**10-13km (~6-8 miles)** — City-level. Good detail for most purposes. - •**25km (~15 miles)** — Regional. Captures large-scale weather patterns but misses local effects. - •**55km (~34 miles)** — Synoptic-scale. Broad patterns only. Individual city conditions are interpolated. 3km models (HRRR, NBM) capture local effects — sea breezes, urban heat islands, terrain-driven variations — that can shift temperatures by several degrees compared to coarser grids. Note HRRR updates hourly at 3km resolution. NBM applies bias correction across multiple model inputs. ECMWF provides an independent global forecast. When these models agree, the signal is consistent. When they diverge, the atmosphere is in a less predictable state. [Market Settlement](https://minutetemp.com/docs/market-settlement)[Reading the Dashboard](https://minutetemp.com/docs/reading-dashboard) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/oracle-scores # Oracle Scores How we measure forecast model accuracy — comparing predicted high/low temperatures against official ASOS observations to rank every model, every day. ## What Is an Oracle Score? An Oracle Score measures how accurate a forecast model's predicted high and low temperatures were compared to what actually happened. Every day, each model produces one or more forecast runs targeting a future date. Once that date passes, we compare every run's prediction against the official observed temperature. The result is a per-model accuracy ranking — lower scores mean more accurate forecasts. Over a multi-day window (e.g., 7 days), you can see which models consistently perform best at a given station. ## Ground Truth: ASOS Observations All scores are measured against observed daily high and low temperatures from **ASOS (Automated Surface Observing System)** stations. These are the same stations that feed NWS official records. We pull the actual high and low in Fahrenheit for the station's local day. If ASOS data is incomplete for a given day (missing high or low), that day is skipped entirely — no partial scores are recorded. ## Three Scoring Modes The Oracle Score widget offers three modes that answer different questions: ### Overall **How accurate is each model across all of its forecast runs?** For a given scored date, we look at _every_ forecast run that produced predictions for that day — regardless of when the run was created. A GFS run from 5 days out and a GFS run from 1 day out both contribute equally. This gives you the model's **overall reliability**. A model that scores well here is consistently accurate at every forecast horizon, not just close-in. ### Day-Ahead **How accurate was the prior day's forecast for the scored day?** For a given scored date, we only consider forecast runs that were fetched during the local day _before_ the scored date. This is the forecast you would have been looking at when making a decision. This is the most relevant mode for **betting and trading**. It answers: "If I trust this model's forecast the day before, how often is it right?" ### Day-Of **How accurate were forecasts fetched during the scored day?** For a given scored date, we only consider forecast runs that were fetched during that same local day. The score is still settled after the full day of observations is available. This mode is useful for same-day decisions and for comparing how models perform once short-range updates are available. Which mode should I use? Use **Day-Ahead** to decide which model to trust for your next bet. Use **Day-Of** for same-day checks once short-range runs are available. Use **Overall** to evaluate which models are fundamentally the best forecasters across all time horizons. ## Explore the Error Distribution Expand a model in the ranking table to see separate high and low error histograms for your selected mode and time window. Errors are **forecast minus observed**: blue bars are too cool, orange bars too warm, and the green band is within half a degree of zero. The bands are 1°F wide, with overflow bars beyond ±10.5°F. P10, median, and P90 summarize the merged forecast-run samples, not averages of daily percentiles. These estimates use band centers; overflow percentiles show bounds instead. Each run has equal weight in a histogram, while MAE and bias give each scored day equal weight. Neither describes the probability of a future temperature outcome. Coverage shows both the sample count per side and the number of scored days with a distribution. Older scores may lack retained forecast samples and are not backfilled with invented data. “Partial coverage” means those days still contribute to the accuracy ranking but not to the histogram or its percentiles. ## How Scores Are Calculated Each model's daily score is computed in three steps: 1. **Extract predicted high/low per run.** For each forecast run, we take the maximum and minimum of all hourly temperature predictions (`temperature_2m_f`) that fall within the station's local day window. 2. **Compare against observed.** For each run, compute the error: `predicted - actual`. The absolute value is the run's error; the signed value is the run's bias. 3. **Average across runs.** All qualifying runs for that model are averaged to produce the day's MAE (Mean Absolute Error) and bias. When you select a multi-day range (e.g., 7 days), the displayed values are the averages of each scored day's MAE and bias across the window. ### Reading the Numbers Bias (High / Low) — Primary The average signed error in degrees Fahrenheit. **Positive (+) means the model runs warm** (over-predicts). **Negative (-) means it runs cold** (under-predicts). This is the most actionable number. If GFS shows a +2° high bias and forecasts 82°F tomorrow, adjust down — expect closer to 80°F. A bot can apply this correction automatically. Color coding: green within ±1°, amber within ±2.5°, red beyond ±2.5°. MAE (High / Low) — Reliability Mean Absolute Error — the average magnitude of the error regardless of direction. Lower is better. In the widget, you can rank by **High MAE** or **Low MAE** depending on which market you are trading. MAE tells you how **reliable** a model is. A model with 0° bias but 5° MAE swings wildly in both directions — you can't correct for it. A model with +2° bias and 2° MAE is consistently warm — very reliable once you adjust. Bias vs MAE for Bots Use **bias** to adjust forecasts (subtract the bias from the model's prediction). Use **High MAE** when trading highs and **Low MAE** when trading lows. Both are available in the API response for programmatic access. ## Scoring Timeline Scoring begins at **12:05 a.m. in each station's local timezone** for the day that just ended. Stations are processed in turn, so scores may appear shortly afterward. Rechecks at 1:05 a.m. and 6:05 a.m. incorporate late observations. Scores can be revised as late data arrives. Scheduled scoring refreshes the cache and sends live updates to the dashboard. Restart catch-up is bounded by the forecast data still retained; it does not reconstruct missing historical samples. Historical scores use your plan's lookback (e.g. 7 / 30 / 90 scored days); Clanker can also request `days=all` for the full retained series. See your [plan tier](https://minutetemp.com/pricing). Free users can view 1 day of scores for public models. ## API Access Oracle Scores are available via the REST API for programmatic access. See the [API Reference](https://minutetemp.com/docs/api) for endpoint details, parameters, and response format. [Reading the Dashboard](https://minutetemp.com/docs/reading-dashboard)[API Reference](https://minutetemp.com/docs/api) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/docs/reading-dashboard # Reading the Dashboard A walkthrough of every element on the city market dashboard — what it shows and how it works. ## Temperature Chart The main chart overlays observed temperature with model forecasts over time. The green line is the latest station observation, and each additional colored line represents a model's prediction for the station temperature. - •Thin **red** and **blue** lines above/below the observation line show the precision window from integer-Celsius reporting - •Lines converging = models agree (high confidence) - •Lines diverging = models disagree (higher uncertainty) - •Hover over the chart to see exact values, including **Actual**, **Min**, and **Max** for observations Toggle individual models on and off to reduce visual noise. See the [forecast models guide](https://minutetemp.com/docs/forecast-models) for details on each model's resolution, update frequency, and coverage. ## Model Pill List Below the chart, color-coded pills show which models are currently displayed. Each pill's color matches its line on the chart. - •Click a pill to toggle that model on or off - •The model selector lets you choose preset groups (e.g., "US models only") - •Your tier determines how many models are available (free: 3, paid: all 19) ## Market Odds Overlay The market odds card shows real-time prices from temperature prediction markets. Prices represent the market's implied probability that the temperature will be above or below specific strike prices. - •**Strike prices** are temperature thresholds (e.g., "Above 75°F") - •**Prices** range from $0 to $1 — a price of $0.70 means the market thinks there's roughly a 70% chance of that outcome - •**Orderbook depth** shows how much liquidity is available at each price level Compare what the models are predicting with what the market is pricing. When they diverge, there may be a trading opportunity — or the market may know something the models don't. ## Market Bracket Bands When market brackets are enabled, colored horizontal bands overlay the chart showing each prediction market bracket's temperature range. Solid 2px border lines mark the exact boundary between adjacent brackets so you can see precisely where one range ends and the next begins. - •**Orange bands** = today's market brackets, **blue bands** = tomorrow's - •**Click a label** on the right side of the chart to toggle that bracket on or off — useful for isolating a specific range - •Hidden brackets show dimmed labels with strikethrough text. Click again to restore, or use the **Show all** button to restore all hidden brackets at once - •Labels show the temperature range and market-implied probability (e.g., "72-76 34%") Toggle bracket bands on or off globally using the chart settings gear icon. The High/Low toggle in the market section controls which bracket set is displayed. ## Market Provider Toggle Switch between the platforms that list the city: Kalshi, Coinbase (which lists Kalshi's markets), Polymarket, Robinhood and IBKR. The page then follows that platform's settlement station and market day. Kalshi, Coinbase and Polymarket show market odds; Robinhood and IBKR show the ForecastEx contracts they list with each strike's last trade and open interest, the station and its Weather Underground settlement data. Not all platforms cover every city. The toggle will only show platforms that have active markets for the city you're viewing. ## Hourly Table The hourly table breaks down each model's temperature prediction hour by hour. This is where you can spot exactly when models diverge. Each row represents one hour. Each column shows a model's forecast for that hour. Look for: - •Hours where all models cluster tightly — high confidence - •Hours where models spread out — the atmosphere is uncertain - •The hour when the high or low is expected to occur — most relevant for settlement ## Observations Ticker The observations ticker shows real-time NWS station readings — the actual measured temperature as reported by the weather station. These are polled every 60 seconds. When the station stops reporting, charts may bridge short gaps visually while the ticker indicates when data is stale. The observed high/low row above the ticker shows both the displayed value and its implied min/max precision bounds so you can quickly gauge rounding risk near a strike. Remember: these observations are [preliminary data](https://minutetemp.com/docs/nws-data). The official record may differ after NWS review. ## Timer Panel The timer shows a countdown to the next data update. Your subscription tier determines update frequency: - •**Free tier:** 3-minute delayed updates - •**Paid tiers:** 1-minute updates ## Resolution & Time Range Two selectors control what you see on the chart: - •**Resolution selector:** Controls how frequently the displayed data updates. Higher resolution means more data points on the chart. Available resolutions depend on your subscription tier. - •**Time range selector:** Zooms the chart in or out on the forecast horizon. Short ranges (6-12 hours) show more detail for imminent weather. Long ranges (3-7 days) show the broader outlook. Shorter time ranges (6-24 hours) show more granular detail. Longer ranges (3-7 days) reveal the broader trend but with less precision per data point. [Forecast Models](https://minutetemp.com/docs/forecast-models) [Back to Docs](https://minutetemp.com/docs) --- Source: https://minutetemp.com/api # minuteTemp API REST and WebSocket APIs for live settlement-station temperatures, forecasts, Kalshi Weather Index minutes and prediction-market metadata, built for trading bots. ## Base URL and authentication REST endpoints live under `https://api.minutetemp.com/api/v1/`. Every request needs an API key from a Starter or Pro plan, sent as an `X-API-Key` header or as `Authorization: Bearer `. Create keys on your profile page after signing up. ``` curl -H "X-API-Key: $MINUTETEMP_API_KEY" https://api.minutetemp.com/api/v1/stations ``` ## Plans and rate limits - **Starter** ($29.99/mo): 15 req/min sustained (about 21,600 requests a day), 30 request burst, WebSocket subscriptions for 6 market + 30 neighbour stations, 7 days of history. - **Pro** ($89.99/mo): 100 req/min sustained (about 144,000 requests a day), 200 request burst, WebSocket subscriptions for 60 market + 300 neighbour stations, 30 days of history. See [pricing](https://minutetemp.com/pricing) for every plan. `GET /api/v1/limits` returns your key's effective limits. ## REST endpoints ### Today - `GET /api/v1/today` — Today's consensus and observed data ### Cities - `GET /api/v1/cities` — List all cities - `GET /api/v1/cities/{slug}` — Get city by slug ### Stations - `GET /api/v1/stations` — List all stations - `GET /api/v1/stations/{station_id}` — Get station details ### Observations - `GET /api/v1/stations/{station_id}/observations/latest` — Get latest observation - `GET /api/v1/stations/{station_id}/observations` — Get historical observations ### Reports - `GET /api/v1/stations/{station_id}/reports/latest` — Get latest reports by type - `GET /api/v1/stations/{station_id}/reports` — Get station reports - `GET /api/v1/stations/{station_id}/reports/history` — List historical station reports ### Weather Events - `GET /api/v1/stations/{station_id}/weather-events` — Get station weather events ### Forecasts - `GET /api/v1/forecasts/models` — List forecast models - `GET /api/v1/stations/{station_id}/forecast` — Get latest forecast - `GET /api/v1/stations/{station_id}/oracle-scores` — Get oracle forecast scores - `GET /api/v1/stations/{station_id}/oracle-score-days` — Get raw daily oracle score components - `GET /api/v1/stations/{station_id}/forecast/runs` — List forecast runs - `GET /api/v1/stations/{station_id}/forecast/runs/{run_id}` — Get forecast run details - `GET /api/v1/data/access` — Get access to forecast data files ### Markets - `GET /api/v1/markets` — List prediction markets - `GET /api/v1/markets/{market_id}` — Get market details - `GET /api/v1/markets/{market_id}/history` — Get market snapshot history ### Hourly Indexes - `GET /api/v1/hourly-indexes` — List hourly index cities - `GET /api/v1/hourly-indexes/{city}` — Get an hourly index city - `GET /api/v1/hourly-indexes/{city}/minutes` — Get hourly index minutes - `GET /api/v1/hourly-indexes/{city}/settlements` — Get hourly index settlements - `GET /api/v1/hourly-indexes/{city}/calibrations` — Get hourly index calibrations - `GET /api/v1/hourly-indexes/{city}/forecast` — Get the hourly index forecast - `GET /api/v1/hourly-indexes/{city}/forecast/history` — Get the stitched hourly index forecast history - `GET /api/v1/hourly-indexes/{city}/match-rate` — Get the settlement match rate - `GET /api/v1/hourly-indexes/{city}/forecast-accuracy` — Get the index forecast accuracy ### Limits - `GET /api/v1/limits` — Get effective limits for current authenticated user ### WebSocket - `POST /api/v1/ws-ticket` — Get a WebSocket ticket ## WebSocket Get a 90-second ticket with `POST /api/v1/ws-ticket`, connect with the WebSocket subprotocols `bearer` and the ticket, then send `subscribe` messages for stations, cities and hourly indexes. Observations, forecast updates, station reports, weather events and market updates arrive as they happen. - `wss://api.minutetemp.com/ws/api/1m` — 1-minute cadence unified channel (Starter/Pro/Clanker plans). - `wss://api.minutetemp.com/ws/api/5m` — Legacy compatibility unified channel (Starter/Pro/Clanker plans). Message shapes and the subscribe protocol are in the [WebSocket docs](https://minutetemp.com/docs/ws). ## Machine-readable specs - OpenAPI 3.1 (REST): [https://api.minutetemp.com/v1/openapi.json](https://api.minutetemp.com/v1/openapi.json) (JSON) and [https://api.minutetemp.com/v1/openapi.yaml](https://api.minutetemp.com/v1/openapi.yaml) (YAML) - AsyncAPI 3.0 (WebSocket): [https://api.minutetemp.com/v1/asyncapi.yaml](https://api.minutetemp.com/v1/asyncapi.yaml) - Site summary for AI assistants: [/llms.txt](https://minutetemp.com/llms.txt) and [/llms-full.txt](https://minutetemp.com/llms-full.txt) The [interactive API reference](https://minutetemp.com/docs/api) renders the OpenAPI spec with request and response examples. [Interactive API reference](https://minutetemp.com/docs/api) [Back to Docs](https://minutetemp.com/docs)